Sol-Ark 15K + SOK batteries = Weird SOC issue

[1201]

I would love to see everything balanced, I was purely trying to say it appears the data I'm seeing from Solar Assistant isn't accurate, thus it's hard for me to base anything off of that.

Clearly balanced cells is important, I totally agree.

If you measure the pack voltages with a multimeter are they identical?

 

[Watts Happening]

I suspect the batteries are daisy chained and not connected to a common bus bar and even still the difference is concerning. 55v is about 95% full while 53v could be 50%.

I suppose you don't believe when I say they are connected to bus bars with equal length cables?

 

[1201]

I suppose you don't believe when I say they are connected to bus bars with equal length cables?

Sorry not calling you a liar, I just don't see how it's possible to be on the same bus bar and have such a variation in voltage. In my mind that's impossible .do you mind posting a pic?

 

[Watts Happening]

Sorry not calling you a liar, I just don't see how it's possible to be on the same bus bar and have such a variation in voltage. In my mind that's impossible .do you mind posting a pic?

Lol, alright... Here's an old picture with an 18kPV and only three batteries... I'm not around the system right now to produce proof of the current configuration, but I'd think we're all capable of understanding what bus bars and battery cables are. I assure you I didn't buy and install bus bars to simply daisy chain my batteries and pretend I hadn't.

 

[G00SE]

I would love to see everything balanced, I was purely trying to say it appears the data I'm seeing from Solar Assistant isn't accurate, thus it's hard for me to base anything off of that.

Clearly balanced cells is important, I totally agree.

And just to clarify. I’m not trying to push back against any of your rationale or points already made. Just thinking out loud I suppose.
I didn’t think you interpreted negatively but I sometimes dwell on nonsense lmao

 

[1201]

Lol, alright... Here's an old picture with an 18kPV and only three batteries... I'm not around the system right now to produce proof of the current configuration, but I'd think we're all capable of understanding what bus bars and battery cables are. I assure you I didn't buy and install bus bars to simply daisy chain my batteries and pretend I hadn't.

View attachment 199407

View attachment 199408

Funny I was just looking at your old threads and had just seen that.

You definitely have bus bars so my mind is fully blown how one battery can be at 55v and another battery on the same bus bar can be at 53v

 

[G00SE]

Funny I was just looking at your old threads and had just seen that.

You definitely have bus bars so my mind is fully blown how one battery can be at 55v and another battery on the same bus bar can be at 53v

Would it be possible that when the packs were commissioned that they were at various states of charge and the voltage cutoffs (particularly upper) never allowed them to balance out and simply drift further apart over time?
If I’m not mistaken he fairly recently upped his voltage limit and previously had it lower
Compound that with individual packs being out of balance and the BMS thresholds may have a hay day?

 

[1201]

Would it be possible that when the packs were commissioned that they were at various states of charge and the voltage cutoffs (particularly upper) never allowed them to balance out and simply drift further apart over time?
If I’m not mistaken he fairly recently upped his voltage limit and previously had it lower
Compound that with individual packs being out of balance and the BMS thresholds may have a hay day?

I'm not sure. I know that when I've introduced new batteries to my system with a bus bar, they charged to the same voltage within minutes

 

[Watts Happening]

And just to clarify. I’m not trying to push back against any of your rationale or points already made. Just thinking out loud I suppose.
I didn’t think you interpreted negatively but I sometimes dwell on nonsense lmao

No not at all, maybe I should clarify my points here.

110% I agree that having cells balanced is key, no question about it. That said, this system has been installed for months and never had any "issues", and by that I mean it's worked all day every day. I'm only now aware of these issues due to the creation of this thread and the pursuit of an additional usable 10% of my battery bank capacity leading me to purchase the SOK cable, which then gave me the ability to use the same cable to connect to Solar Assistant when not being used for programming.

After connecting that, I noticed "things not making sense". We all make assumptions, I'm guilty of that. I assumed that the (3) batteries I initially had, all purchased within a relatively short period of time (I don't recall how far apart) would have had the same set points. I've not heard or seen anyone updating SOK firmware, nor have I seen a firmware update talked about on the Current Connected site. With that, I never even considered that they would have differing settings for voltage set points, regardless of firmware. Probably a bad assumption.

I also hadn't monitored cell level parameters because again, I assumed the BMS was doing its job.

I then assumed that the battery reporting to be at 95% had some form of an issue, I never imagined the BMS itself would report a different value locally.

As soon as I realized the value locally and on Solar Assistant wasn't the same, I then thought, well, ANY of the data could be inaccurate, cell voltages etc.

The facts I know to be true are this: They are all SOK 48v batteries from Current Connected, they are all in closed loop communication with the Sol-Ark 15k, I have not made any changes to the settings via SOKTools, they share a common bus bar and equal length battery cables. I would imagine they would balance pretty accurately over time.

I'd like to A) understand where the inaccuracies in reporting through Solar Assistant derive from (doesn't matter, just want to fix it), this way I can monitor data and believe what I'm seeing, which will help me diagnose future issues as I don't particularly want to climb up on my gun safe every time to check things, B) get the cells as balanced as possible (I assume a step in this process is to get all batteries to identical parameters and/or the same firmware) and once those two are accomplished I'd like to C) determine a way to get the BMS to listen to the value input in the "SOC Low Alarm %" field.

 

[G00SE]

As soon as I realized the value locally and on Solar Assistant wasn't the same, I then thought, well, ANY of the data could be inaccurate, cell voltages etc.

Interesting, I hadn’t even considered that but makes sense to the rest of your considerations!

 

[Watts Happening]

Interesting, I hadn’t even considered that but makes sense to the rest of your considerations!

I will try to find the time this evening to connect SOKTools to each battery and log current settings and current cell voltages, then compare to what I'm seeing in Solar Assistant. At the end of the day Solar Assistant is just a convenience, so long as things at the pack/cell level are actually where they are supposed to be, the rest is just tinkering and 'fun'.

For all I know, the BMS's are doing exactly what they are supposed to and all cells are within .000001v right now.

 

[42OhmsPA]

I'm really curious if the values would match in an open loop setup with the batteries connected directly to SA.

They are currently being read from the solark and passed to SA, correct?
Possibly one of them is averaging the values?

 

[Watts Happening]

I'm really curious if the values would match in an open loop setup with the batteries connected directly to SA.

They are currently being read from the solark and passed to SA, correct?
Possibly one of them is averaging the values?

Per the settings in Solar Assistant no, they should not be coming from Sol-Ark. I have inverter set to one of my two USB ports and battery connected to the other (Solar Assistant checks the connections and reports if they are 'connected').

Up until receiving that cable from Current Connected I WAS using inverter values in Solar Assistant, it was just seeing what was passed from Sol-Ark which equated to an averaged value of the connected battery packs.

Now that I have the packs connected via a USB adapter directly to Solar Assistant I can see much more granular information, albeit I'm not convinced it's all 100% accurate.

I will however remove the communication from Sol-Ark and see if it magically makes any changes on the Solar Assistant side of things.

 

[42OhmsPA]

Per the settings in Solar Assistant no, they should not be coming from Sol-Ark. I have inverter set to one of my two USB ports and battery connected to the other (Solar Assistant checks the connections and reports if they are 'connected').

Up until receiving that cable from Current Connected I WAS using inverter values in Solar Assistant, it was just seeing what was passed from Sol-Ark which equated to an averaged value of the connected battery packs.

Now that I have the packs connected via a USB adapter directly to Solar Assistant I can see much more granular information, albeit I'm not convinced it's all 100% accurate.

I will however remove the communication from Sol-Ark and see if it magically makes any changes on the Solar Assistant side of things.

I doubt it will.

I should have stated long ago that I've never used closed loop and all of my thoughts are from what I've learned in an open loop setup, I also have a different brand BMS.

I was experiencing random communication errors until I jumped to the beta version and kept up with updates.
Have you noticed any errors in the SA settings tab?

You can see a log in the USB details section as well.

 

[Watts Happening]

No errors shown, USB settings look accurate to me, I’ve played with it all many times this week.

 

[42OhmsPA]

Looking forward to seeing what the experts at CC come up with.
Fortunately it's still up and running.
I'm guessing identical firmware versions or adjusting all BMS settings to be the same.

 

[Watts Happening]

After going through all batteries with SOKTools this morning I've determined the following information (I have screenshots of everything, but it's just a lot to look at on a forum. Notice the only battery with differing values is #2, which happens to be the battery with the oldest firmware. It's also NOT the battery giving me "trouble" from a Solar Assistant reporting perspective. I've tried to change the settings to match the others, but the software won't let me "write" to the battery, I'm sure there's a reason for that, but I'd love to get it to the same spec.

Firmware Version
1 - 3.01
2 - 1.04
3 - 3.01
4 - 4.07
5 - 4.07

Voltages
1 - 55.031
2 - 56.897
3 - 56.701
4 - 53.978
5 - 54.342

High Cell/Low Cell
1 - 3446/3430
2 - 3562/3541
3 - 3556/3486
4 - 3442/3345
5 - 3465/3352

Warnings
1 - None
2 - None
3 - None
4 - Cell OVP
5 - Cell OVP

Cell OVP Release
1 - 3.38
2 - 3.50
3 - 3.38
4 - 3.38
5 - 3.38

SCP Delay Time
1 - 300
2 - 350
3 - 300
4 - 300
5 - 300

Balance Threshold V
1 - 3.50
2 - 3.45
3 - 3.50
4 - 3.50
5 - 3.50

Balance Cell mV
1 - 30
2 - 20
3 - 30
4 - 30
5 - 30

Pack Full Charge Voltage
1 - 56.00
2 - 57.60
3 - 56.00
4 - 56.00
5 - 56.00

Now, one thing I did notice that may be of value is that pack #3 (the one that refuses to show above 95% in Solar Assistant) is the only pack not showing a 100% state of health. In SOK tools you'll notice it is listed as 95%, so, is there some correlation between "health" and max percentage through Solar Assistant? Maybe that value gets reported in such a way that a battery with a sub-100% state of health is no longer able to reach 100% SOC and is capped at whatever the SOH value is? It seems rather coincidental that the value matches identically, but who knows.

And to reiterate, the battery reports 100% SOC locally on the BMS, but when reported through Solar Assistant it maxes out at 95%




I now feel like I at least have some data for them, I would however like to figure out how to get them all balanced as accurately as possible so they're all "in line". @CCJ

 

[G00SE]

I believe @marionw has done a lot of digging into the firmware iterations. Hoping they’ll chime in with potentially useful information

Edit. I see they’ve already contributed in here. I saw a separate post in Kenny’s thread this morning which made me think of it

 

[Watts Happening]

I believe @marionw has done a lot of digging into the firmware iterations. Hoping they’ll chime in with potentially useful information

Edit. I see they’ve already contributed in here. I saw a separate post in Kenny’s thread this morning which made me think of it

I’m not sure what happens behind the scenes in the firmware, but the changes seem minimal at least via the SOKTools software

 

[1201]

@Watts Happening have you by chance already measured the reported voltages with a multimeter?

 

[Watts Happening]

@Watts Happening have you by chance already measured the reported voltages with a multimeter?

I haven’t yet, family in town and a 3 year old only left so much time this morning before leaving for breakfast.

What I DID do though before leaving was set the Sol-Ark to charge batteries to 100%, also set the Chargeverter to charge at a low amperage and turned battery #2 off.

I did this because #2 was sitting at the highest voltage and my thinking was removing it from the equation would give the other batteries a chance to balance. Whether this is true or not, I’ll find out later.

 

[marionw]

After going through all batteries with SOKTools this morning I've determined the following information (I have screenshots of everything, but it's just a lot to look at on a forum. Notice the only battery with differing values is #2, which happens to be the battery with the oldest firmware. It's also NOT the battery giving me "trouble" from a Solar Assistant reporting perspective. I've tried to change the settings to match the others, but the software won't let me "write" to the battery, I'm sure there's a reason for that, but I'd love to get it to the same spec.

Firmware Version
1 - 3.01
2 - 1.04
3 - 3.01
4 - 4.07
5 - 4.07

Voltages
1 - 55.031
2 - 56.897
3 - 56.701
4 - 53.978
5 - 54.342

High Cell/Low Cell
1 - 3446/3430
2 - 3562/3541
3 - 3556/3486
4 - 3442/3345
5 - 3465/3352

Warnings
1 - None
2 - None
3 - None
4 - Cell OVP
5 - Cell OVP

Cell OVP Release
1 - 3.38
2 - 3.50
3 - 3.38
4 - 3.38
5 - 3.38

SCP Delay Time
1 - 300
2 - 350
3 - 300
4 - 300
5 - 300

Balance Threshold V
1 - 3.50
2 - 3.45
3 - 3.50
4 - 3.50
5 - 3.50

Balance Cell mV
1 - 30
2 - 20
3 - 30
4 - 30
5 - 30

Pack Full Charge Voltage
1 - 56.00
2 - 57.60
3 - 56.00
4 - 56.00
5 - 56.00

Now, one thing I did notice that may be of value is that pack #3 (the one that refuses to show above 95% in Solar Assistant) is the only pack not showing a 100% state of health. In SOK tools you'll notice it is listed as 95%, so, is there some correlation between "health" and max percentage through Solar Assistant? Maybe that value gets reported in such a way that a battery with a sub-100% state of health is no longer able to reach 100% SOC and is capped at whatever the SOH value is? It seems rather coincidental that the value matches identically, but who knows.

And to reiterate, the battery reports 100% SOC locally on the BMS, but when reported through Solar Assistant it maxes out at 95%


View attachment 199535

I now feel like I at least have some data for them, I would however like to figure out how to get them all balanced as accurately as possible so they're all "in line". @CCJ

Try paceadmin for the Administrator Password.

The SOKTools I am using I downloaded from https://www.currentconnected.com/sk48v100-owner-operators-manual/#commintro

 

[Kenny_]

I haven’t yet, family in town and a 3 year old only left so much time this morning before leaving for breakfast.

What I DID do though before leaving was set the Sol-Ark to charge batteries to 100%, also set the Chargeverter to charge at a low amperage and turned battery #2 off.

I did this because #2 was sitting at the highest voltage and my thinking was removing it from the equation would give the other batteries a chance to balance. Whether this is true or not, I’ll find out later.

I've done that a couple of times in order to try and balance out the batteries and it did help, but the thing that actually worked was to get the inverter out of the mix and let the battery BMS handle things. I went to Open Loop/User mode and that has made a huge difference! No weird alarms, not batteries acting weird etc. etc.

 

[Kenny_]

Try paceadmin for the Administrator Password.

LOL! I was just trying to remember that this morning after not having to use it for a couple of months.... wrote it down after I finally remembered.....thought I was going to have to ask you for it.

 

[Kenny_]

The only way I know to get the BMS to reset what 100% SOC actually is is to get the battery cells to the Balance voltage and hold it there. If you are running in "closed loop"

I'm relatively confident that this was my ongoing problem with certain batteries not 'fully' charging. I still think the inverter and the batteries were fighting one-another.

Once I got the inverter out of the way by going to user mode and open loop everything has been fine.